523
5 CONCLUSIONS
In the paper the results of preliminary analysis on dynamic excitation to Haul Truck
operated in deep underground mine have been presented. It was shown that signals are
extremely nonstationary, depend on several groups of factors. To deduce from the data,
the analysis should be done according to the context, i.e. before interpretation, one should
make sure that the impact from other factors is minimized. The main purpose of this
work is to examine the external excitation used in the machine, which is related to the
uneven surface. External excitation might be related to the payload (weight of material
carried by the truck box), several operational parameters (speed, rpm, gear), environmental factors (mud, water) as well of the geometry of the corridor (elevations, falls, turns,
intersections) and finally skills of the operator. In order to really investigate influence of
road surface and simplify the analysis, we introduced gear number context—only data
acquired during driving on gear number 2 have been analyzed. Gear = 2 is the most used
gear for “stationary” driving. It has been found that basic statistical analysis is not able
to recognize differences between vibration for the loaded and unloaded machine. We have
proposed additional criteria for segmentation, namely the speed, the fuel consumption
and the EngineRPM speed.
Detailed analysis of acceleration in three XYZ directions allows identification of some
typical behavior of an operator/machine. We have confirmed, that close to mining screen,
vibration registered by sensors are smaller (in sense of variance of vibration), while close to
mining face the are much intense. It is related to road quality (closer to mining face—worse
road quality). In this paper, we completely ignore highly transient regimes when the operator
is changing the gear during acceleration or deceleration. This issue will be considered in the
future research.
Figure 10. vibration data for pre-segmented measurement 2 and measurement 4 (after unloading)
a) First segment with good road (close to unloading point), 2nd-4th segments rough surface
b) First segment with good road (close to unloading point), 2nd-3rd segments rough surface.
Figure 11. The hydraulic pressure signal for loaded and unloaded machine operated under Gear 2.
2.5
- 1.5
-o.5
-2
·2.5
91 .291 .491.691.8
T ime(min)
Time (min]
,.
o.a -measurement 1
02
..
1.5
L5
0.5
~
< -5
·1.5
-o.5
"0
'"
...
216.8 217 21 7.2217.4
Time(mln)
T ime [m in)
Time (min]
measurement 2
..
•I
..
........
measurement 3
2.5
if~
·1.5
·2.5 218 218.5 219
Trne(minJ
measurem.ent 4
. .
. ~
..
5 CONCLUSIONS
In the paper the results of preliminary analysis on dynamic excitation to Haul Truck
operated in deep underground mine have been presented. It was shown that signals are
extremely nonstationary, depend on several groups of factors. To deduce from the data,
the analysis should be done according to the context, i.e. before interpretation, one should
make sure that the impact from other factors is minimized. The main purpose of this
work is to examine the external excitation used in the machine, which is related to the
uneven surface. External excitation might be related to the payload (weight of material
carried by the truck box), several operational parameters (speed, rpm, gear), environmental factors (mud, water) as well of the geometry of the corridor (elevations, falls, turns,
intersections) and finally skills of the operator. In order to really investigate influence of
road surface and simplify the analysis, we introduced gear number context—only data
acquired during driving on gear number 2 have been analyzed. Gear = 2 is the most used
gear for “stationary” driving. It has been found that basic statistical analysis is not able
to recognize differences between vibration for the loaded and unloaded machine. We have
proposed additional criteria for segmentation, namely the speed, the fuel consumption
and the EngineRPM speed.
Detailed analysis of acceleration in three XYZ directions allows identification of some
typical behavior of an operator/machine. We have confirmed, that close to mining screen,
vibration registered by sensors are smaller (in sense of variance of vibration), while close to
mining face the are much intense. It is related to road quality (closer to mining face—worse
road quality). In this paper, we completely ignore highly transient regimes when the operator
is changing the gear during acceleration or deceleration. This issue will be considered in the
future research.
Figure 10. vibration data for pre-segmented measurement 2 and measurement 4 (after unloading)
a) First segment with good road (close to unloading point), 2nd-4th segments rough surface
b) First segment with good road (close to unloading point), 2nd-3rd segments rough surface.
Figure 11. The hydraulic pressure signal for loaded and unloaded machine operated under Gear 2.
2.5
- 1.5
-o.5
-2
·2.5
91 .291 .491.691.8
T ime(min)
Time (min]
,.
o.a -measurement 1
02
..
1.5
L5
0.5
~
< -5
·1.5
-o.5
"0
'"
...
216.8 217 21 7.2217.4
Time(mln)
T ime [m in)
Time (min]
measurement 2
..
•I
..
........
measurement 3
2.5
if~
·1.5
·2.5 218 218.5 219
Trne(minJ
measurem.ent 4
. .
. ~
..